Data assimilation an overview
1 / 39

Data Assimilation – An Overview - PowerPoint PPT Presentation

  • Uploaded on

Data Assimilation – An Overview. Outline: What is data assimilation? What types of data might we want to assimilate? Optimum Interpolation Simple examples of OI Variational Analysis a generalization of OI Bayesian methods (simple example) Integration Common problems

I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.
Download Presentation

PowerPoint Slideshow about ' Data Assimilation – An Overview' - dante

An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.

- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript
Data assimilation an overview
Data Assimilation – An Overview


  • What is data assimilation?

  • What types of data might we want to assimilate?

  • Optimum Interpolation

  • Simple examples of OI

  • Variational Analysis a generalization of OI

  • Bayesian methods (simple example)

  • Integration

  • Common problems

  • Rarely is one method used exclusively

  • G. Levy – Data Assimilation/Forecasting


    • Lao Tzu, Chinese Philosopher

      “He who knows does not predict.

      He who predicts does not know.”

    G. Levy – Data Assimilation/Forecasting

    The purpose of data assimilation
    The Purpose of Data Assimilation

    • To combine measurements and observations with our knowledge of statistics and the physical system’s behavior as modeled to produce a “best” estimate of current conditions.

    • The analysis has great diagnostic value and is the basis for numerical prediction.

    • It allows to control of model error growth

    G. Levy – Data Assimilation/Forecasting

    Why use a forecast model to generate background field
    Why use a (forecast) model to generate background field?

    • Dynamical consistency between mass and motion

    • Advection of information into data-sparse regions

    • Improvement over persistence and climatology

    • Temporal continuity

    G. Levy – Data Assimilation/Forecasting

    Necessary conditions to predict
    Necessary Conditions to Predict

    G. Levy – Data Assimilation/Forecasting

    The data assimilation cycle
    The Data Assimilation Cycle

    • Quality Control

    • Objective Analysis (estimating parameters at points where no observations are available, usually on a grid)

    • Initialization

    • Short forecast prepared as next background field

      Discussion break: what are unique challenges and advantages of assimilating remote-sensor data?

    G. Levy – Data Assimilation/Forecasting

    G. Levy – Data Assimilation/Forecasting

    Data assimilation cycle
    Data Assimilation Cycle variable tradeoffs

    • Data Assimilation: the fitting of a prediction model to observed data (reality check, Daley text).

    G. Levy – Data Assimilation/Forecasting

    Objective analysis methods
    Objective Analysis Methods variable tradeoffs

    • Surface Fitting

      • ordinary/weighted least squares fitting

      • principal components (EOF) improve numerical stability

      • spline functions control smoothness

      • cross-validation controls smoothness and sharpness of filters

    • Empirical linear interpolation

      • successive correction techniques

    • Optimal Interpolation

      Above still widely used in research, less so in operations

    • Bayesian Approach

    • Newer hybrid techniques (variants of OI)

      • Adaptive filtering (Kalman filter)

      • Spectral Statistical Interpolation

      • Nudging and 4D Variational analysis

    G. Levy – Data Assimilation/Forecasting

    A simple problem
    A simple problem variable tradeoffs

    • Estimate an unknown quantity x from two collocated measurements y1 & y2 subject to errors

    G. Levy – Data Assimilation/Forecasting

    Simple problem continued
    Simple Problem (continued) variable tradeoffs

    • Constraints imply:

      Solution for a1, a2 is a function of 2

    G. Levy – Data Assimilation/Forecasting

    Equivalent problem
    Equivalent Problem variable tradeoffs

    • Find an estimate  of x that is close to the observations.

    • Do this by minimizing the “distance” (penalty/cost function) between  and the observations

    G. Levy – Data Assimilation/Forecasting

    Optimum interpolation
    Optimum Interpolation variable tradeoffs

    • A generalization of the previous problem in that that the observations may or may not coincide with points where estimates are desired.

    • Interpret Obs. broadly to also include model predictions, so we have:

      • An a priori estimate on a grid provided by model

      • measurements

        For now we do not distinguish between the two kinds of observations.

    G. Levy – Data Assimilation/Forecasting

    Optimum interpolation1
    Optimum Interpolation variable tradeoffs










    X- true state on grid; y - observations;  - error in going from x to y

    G. Levy – Data Assimilation/Forecasting

    Optimum interpolation formulation
    Optimum Interpolation Formulation variable tradeoffs

    G. Levy – Data Assimilation/Forecasting

    Oi continued ignore details
    OI (continued, ignore details) variable tradeoffs

    G. Levy – Data Assimilation/Forecasting

    Oi continued
    OI (continued) variable tradeoffs

    • The solutions to A & P involve only , the error covariance matrix (non-trivial), and the H matrix. This is the basis of OI.

    G. Levy – Data Assimilation/Forecasting

    Variational data assimilation methods
    Variational Data Assimilation Methods variable tradeoffs



    A generalization of OI that

    addresses incompatibility between

    model and measurement space








    Corrected model


    G. Levy – Data Assimilation/Forecasting

    Variational assimilation
    Variational Assimilation variable tradeoffs

    G. Levy – Data Assimilation/Forecasting

    Variational analysis summary
    Variational Analysis Summary variable tradeoffs

    • A vector of observations is related to the true state through a function (possibly non-linear) that includes:

      • Interpolation of variables to observation points

      • Conversion of state to observed variables

    • Observations differ from true value due to:

      • Measurement and background (model) errors

      • Errors of representativeness (sub-grid variability)

    • An analyzed state close to observations and model is found by minimizing a cost function that separates the different errors and includes their covariances.

    G. Levy – Data Assimilation/Forecasting

    Summary of oi

    A widely used statistical approach variable tradeoffs

    A good framework for multivariate analysis that allows incorporation of physical constraints

    Weights determined as function of distribution and accuracy of data and models (RS discussion)

    Can be computationally demanding

    Scale-dependent correlation models require long history for accurate estimates of empirical coefficients (RS discussion)

    Underperforms in extreme events

    Summary of OI

    G. Levy – Data Assimilation/Forecasting

    Bayesian Methods variable tradeoffs

    General characteristics

    • The Bayesian approach allows one to combine information from different sources to estimate unknown parameters.

    • Basic principles:

    • - Both data and external information (prior) are used.

    • Computations are based on the Bayes theorem.

    • Parameters are defined as random variables.

    G. Levy – Data Assimilation/Forecasting

    Bayesian theory historical perspective
    Bayesian Theory - historical perspective variable tradeoffs

    Bayes, T. 1763. An essay towards solving a problem in the doctrine of chances. Philos. Trans. Roy. Soc. London, 53, 370-418.

    This 247 y.o. paper is the basis of the cutting edge methodology of data assimilation and forecasting (as well as other fields).

    G. Levy – Data Assimilation/Forecasting

    Basic Probability Notions variable tradeoffs

    • Basic probability notions needed for applying Bayesian methods:

      • Joint probability.

      • Conditional probability.

      • Marginal probability.

      • Bayes theorem.

    • Consider two random variables A and B representing two possible events.

    G. Levy – Data Assimilation/Forecasting

    Marginal (prior) probability variable tradeoffs

    • A and B are mutually exclussive events.

    • Marginal probability of A = probability of event A

    • Marginal probability of B = probability of event B

    • Notation: P(A), P(B).

    G. Levy – Data Assimilation/Forecasting

    Joint probability variable tradeoffs

    • Joint probability = probability of event A and event B.

    • Notation: P(AB) or P(A, B).

    Conditional probability

    • Conditional probability = probability of event B given event A.

    • Notation: P(B | A).

    G. Levy – Data Assimilation/Forecasting

    Bayes theorem variable tradeoffs

    Bayes’ theorem allows one to relate P(B | A) to P(B) and P(A|B).

    P(B | A) = P(A | B) P(B) / P(A)

    This theorem can be used to calculate the probability of event B given event A.

    In practice, A is an observation and B is an unknown quantity of interest.

    G. Levy – Data Assimilation/Forecasting

    How to use bayes theorem
    How to use Bayes theorem? variable tradeoffs

    G. Levy – Data Assimilation/Forecasting

    Bayes Theorem Example variable tradeoffs

    A planned IO & SCS workshop at the SCSIO Nov. 17 -19. Climatology is of six days of rain in Guanzhou in November. Long term model forecast is of no rain for the 17th. When it actually is sunny, it is correctly forecasted 85% of the time.

    Probability that it will rain?

    A = weather in Day Bay on Nov. 30 (A1: «it rains », A2: « it is sunny»).

    B = The model predicts sunny weather

    G. Levy – Data Assimilation/Forecasting

    Bayes Theorem Example variable tradeoffs

    In terms of probabilities, we know:

    P( A1 ) = 6/30 =0.2 [It rains 6 days in November.]

    P( A2 ) = 24/30 = 0.8 [It does not rain 24 days in November.]

    P( B | A1 ) = 0.15 [When it rains, model predicts sun 15% of the time.]

    P( B | A2 ) = 0.85 [When it doesn’t rain, the model correctly predicts sun 85% of the time.]

    We want to know P( A1 | B ), the probability it will rain, given

    a sunny forecast.

    P(A1 | B) = P(A1) P(B | A1) / [P(A1) P(B | A1) + P(A2) P(B | A2)]

    P(A1 | B) = (0.2) (0.15)/[(0.2) (0.15) + (0.8) (0.85)]

    P(A1 | B) = 0.042

    G. Levy – Data Assimilation/Forecasting

    Forecasting exercise
    Forecasting Exercise variable tradeoffs

    • Produce a probability forecast for Monday

    • Propose a metric to evaluate your forecast

      Possible sources for climatological and model forecasts for Taiwan:

    G. Levy – Data Assimilation/Forecasting

    Evaluating scoring probability forcast
    Evaluating (scoring) Probability forcast variable tradeoffs

    Compare the forecast probability of an event pi to the observed occurrence oi, which has a value of 1 if the event occurred and 0 if it did not occur.

    The BS measures the mean squared probability error

    over N events (RPS can be used for a multi-category

    probabilistic forecast.

    G. Levy – Data Assimilation/Forecasting

    Bayesian data assimilation variable tradeoffs

    What is the multidimensional probability of a particular state given a numerical forecast (first guess) and a set of observations

    : vector of model parameters.

    y: vector of observations

    P(): prior distribution of the parameter values.

    P(y|): likelihood function.

    P(| y): posterior distribution of the parameter values.

    often difficult to compute

    G. Levy – Data Assimilation/Forecasting

    Hybrid methods and recent trends
    Hybrid Methods and Recent Trends variable tradeoffs

    • Kalman Filter: a generalization of OI where the error statistics evolve through (i) a linear prediction model, and (ii) the observations.

    • Spectral Statistical Interpolation (SSI): an extension of OI to 3-D in spectral (global) space.

    • Adjoint method (4DVAR): Extension of Bayesian approach to 4D. Given a succession of observations over time, find a succession of model states

    • Nudging: A model is “nudged” towards observations by including a time dependent term.

    G. Levy – Data Assimilation/Forecasting

    Common problems in data assimilation
    Common Problems in Data Assimilation variable tradeoffs

    • The optimal estimate is not a realizable state

      • Will produce noise if used to initialize

    • The observed variables do not match the model variables

    • The distribution of observations is highly non-uniform

      • Engineering and science specs clash

      • Over and under sampling due to orbit

    • The observations are subject to errors

      • How closely should one fit the observations

      • Establishing errors and error covariances is not trivial…

    • Very large dimensionality of problem

    • Nonlinear transformations and functions

    G. Levy – Data Assimilation/Forecasting

    References textbooks
    References (textbooks) variable tradeoffs

    • Atmospheric Data Analysis by Roger Daley. Published by Cambridge University Press, 1992. ISBN: 0521458250, 472 pp.

    • Atmospheric Modeling, Data Assimilation, and Predictability by Eugenia Kalnay. Published by Cambridge University Press, 2003. ISBN 0521796296, 9780521796293. 341 pages

    G. Levy – Data Assimilation/Forecasting